Characterizing Mechanical Properties of Primary Cell Wall in Living Plant Organs Using Atomic Force Microscopy

Summary

Studies of cell wall biomechanics are essential for understanding plant growth and morphogenesis. The following protocol is proposed to investigate thin primary cell walls in the internal tissues of young plant organs using atomic force microscopy.

Abstract

The mechanical properties of the primary cell walls determine the direction and rate of plant cell growth and, therefore, the future size and shape of the plant. Many sophisticated techniques have been developed to measure these properties; however, atomic force microscopy (AFM) remains the most convenient for studying cell wall elasticity at the cellular level. One of the most important limitations of this technique has been that only superficial or isolated living cells can be studied. Here, the use of atomic force microscopy to investigate the mechanical properties of primary cell walls belonging to the internal tissues of a plant body is presented. This protocol describes measurements of the apparent Young's modulus of cell walls in roots, but the method can also be applied to other plant organs. The measurements are performed on vibratome-derived sections of plant material in a liquid cell, which allows (i) avoiding the use of plasmolyzing solutions or sample impregnation with wax or resin, (ii) making the experiments fast, and (iii) preventing dehydration of the sample. Both anticlinal and periclinal cell walls can be studied, depending on how the specimen was sectioned. Differences in the mechanical properties of different tissues can be investigated in a single section. The protocol describes the principles of study planning, issues with specimen preparation and measurements, as well as the method of selecting force-deformation curves to avoid the influence of topography on the obtained values of elastic modulus. The method is not limited by sample size but is sensitive to cell size (i.e., cells with a large lumen are difficult to examine).

Introduction

The mechanical properties of the plant cell wall determine the shape of the cell and its ability to grow. For example, the growing tip of the pollen tube is softer than the non-growing parts of the same tube¹. The primordia formation on Arabidopsis meristem is preceded by a local decrease in cell wall stiffness at the site of the future primordium^2,3. The cell walls of Arabidopsis hypocotyl, which are parallel to the main growth axis and grow faster, are softer than those that are perpendicular to this axis and grow slower^4,5. In the maize root, the transition of cells from division to elongation was accompanied by a decrease in elastic moduli in all tissues of the root. The moduli remained low in the elongation zone and increased in the late elongation zone⁶.

Despite the availability of various methods, the large arrays of biochemical and genetic information on cell wall biology obtained annually are rarely compared with the mechanical properties of cell walls. For example, mutants on cell wall-related genes often have altered growth and development^4,7,8, but are rarely described in terms of biomechanics. One of the reasons for this is the difficulty of conducting measurements at the cellular and subcellular levels. Atomic force microscopy (AFM) is currently the primary approach for such analyses⁹.

In recent years, numerous AFM-based studies on plant cell wall biomechanics have been carried out. The mechanical properties of cell walls of the outer tissues of Arabidopsis^{2,3,4,5,10,11} and onion¹², as well as of cultured cells^13,14,15, have been investigated. However, the superficial cells of a plant may have cell walls whose mechanical properties differ from those of the inner tissues⁶. In addition, plant cells are pressurized by turgor which makes them stiffer. To get rid of the influence of turgor pressure, researchers have to use plasmolyzing solutions^{2,3,4,5,10,11} or decompose the values obtained into turgor and cell wall contributions¹². The first approach leads to sample dehydration and changes the thickness and properties of the cell wall¹⁶, while the second approach requires additional measurements and complicated mathematics, and applies only to cells of relatively simple shape¹². The cell wall properties of internal tissues can be evaluated on cryosections¹⁷ or sections of plant material impregnated with resin⁸. However, both methods involve dehydration and/or impregnation of samples, which inevitably leads to changes in properties. The properties of isolated or cultured cells are difficult to relate to the physiology of the whole plant. Both cultivation and isolation of plant cells can affect the mechanical properties of their cell walls.

The method presented here complements the aforementioned approaches. Using it, the primary cell walls of any tissue and at any stage of plant development can be examined. Sectioning and AFM observations were performed in liquid which avoids sample dehydration. The problem of turgor was solved as the cells are cut. The protocol describes work with maize and rye roots, but any other sample can be examined if it is suitable for vibratome sectioning.

The AFM studies described here were performed using the force-volume technique. Different instruments use different names for this method. However, the basic principle is the same; a force-volume map of the sample is obtained by a sinusoidal or triangular motion of the cantilever (or sample) to achieve a certain loading force at each analyzed point, while recording the cantilever deflection¹⁸. The result combines a topographic image of the surface and the array of force-distance curves. Each curve is used to calculate the deformation, stiffness, Young's modulus, adhesion, and energy dissipation at a specific point. Similar data can be obtained by point-by-point force-spectroscopy after scanning in contact mode¹⁹, although it is more time-consuming.

Protocol

1. Sample preparation for AFM measurements

1. Plant material: Sterilize the seeds of maize (Zea mays L.) and rye (Secale cereale L.) with a 0.35% NaOCl solution for 10 min, wash 3x with distilled water, and then grow hydroponically in the dark at 27 °C for 4 days and 2 days, respectively. Primary roots were used for the experiment.
2. Preparation of solutions and sample for vibratome sectioning
   1. Prepare agarose solution for root embedding by dissolving 3% (w/w) low-melting-point agarose in water using a microwave oven.
      NOTE: All experiments were performed in water. If buffer or any other type of medium is required for the study, it is better to use the same medium for agarose preparation, during sectioning, and in the liquid cell of the AFM to avoid damaging the sample.
   2. Pour a 4 mm layer of melted 3% agarose on the bottom of the Petri dish (diameter = 35 mm), and let it cool slightly to prevent thermal damage to the sample.
   3. Place three or four small pieces (about 5 mm long) of the studied plant organ horizontally on the agarose.
      NOTE: Samples should be half-immersed but not submerged in the agarose layer. If the sample is not immersed, it may break out of the agarose during vibratome sectioning. If the sample sinks to the bottom, it will be too close to the edge of the section and become inconvenient for further steps.
   4. When a thin, semi-solid film appears on top of the first layer of agarose (30-60 s), carefully pour a second layer on top.
      NOTE: The bottom layer of agarose should not completely solidify. Otherwise, the two layers may separate from each other during further work.
   5. After the agarose is completely solidified, cut out the block containing the specimen. Shape the block into a hexagonal truncated pyramid to ensure its stability during further sectioning (Figure 1A).
      NOTE: The sample can be oriented vertically or horizontally within this block depending on the type of section required. When preparing the block, leave 2-3 mm of agarose around the specimen. In this case, the sample section will be retained by the layer of 3% agarose, which will facilitate its further immobilization (Figure 1B).
3. Sectioning the sample with the vibratome
   1. Glue the block to the vibratome stage with cyanoacrylate adhesive. Place the stage in the vibratome so that one of the pyramid corners faces the blade of the vibratome. Pour water into the vibratome bath.
   2. Set the sectioning parameters (section thickness, blade speed, and vibration frequency), and cut the sample.
      NOTE: Use the section thicknesses between 200 and 400 µm. A section that is too thin can introduce errors in the evaluation of mechanical properties, while a section that is too thick is prone to breakage. A blade speed of 1.3 mm·s^-1 and an oscillation frequency of 90 Hz were used for rye and maize roots.
   3. Using a fine brush, move the section from the water bath onto a glass slide, and place a drop of water on the section to prevent it from drying out. Check the quality of the section under a light microscope.
      NOTE: Oblique sections cannot be used to measure the modulus of elasticity. Cell walls must be perpendicular to the section plane, otherwise they may bend or buckle under the AFM tip.
4. Immobilizing the section for AFM measurements (Figure 1B)
   1. Pour a layer of molten 1% (w/w) agarose (~1 mL) on the bottom of the Petri dish cap using a pipette.
      NOTE: Make sure that the sides of the Petri dish cap do not prevent the tip from approaching the sample. The agarose layer should be 1 mm thick and should cover the entire bottom of the cap.
   2. After the 1% agarose has solidified, remove excess water from the section by bringing filter paper to its edge.
      NOTE: Do not touch the plant section with the paper to avoid damage and complete drying of the sample.
   3. Carefully transfer the section from the slide to the center of the Petri dish cap using a brush. Using a 20 µL pipette, carefully add 1% agarose around the section (Figure 1B).
      NOTE: The 1% agarose should not get on the specimen itself. It should only cover the edges of the 3% agarose layer that retains the plant specimen. The 1% agarose should form small knolls on the edges of this 3% agarose layer. Large knolls may prevent the cantilever from approaching the sample.
   4. Pour the water or other solution to be used for the AFM into the Petri dish cap with the immobilized section.

2. AFM preparation and calibration

NOTE: The force-volume method of AFM generates a spatially resolved array of force-distance curves obtained at each point of the area studied. Obtain all parameters for the force-volume mode (cantilever stiffness, IOS, etc.) in contact mode. Similar procedures for instruments from other manufacturers have been described previously^10,20.

1. Select the appropriate type of cantilever.
   NOTE: The cantilever stiffness must be comparable to the specimen stiffness²¹. A cantilever that is much stiffer than the specimen will not deflect much, while a cantilever that is too soft will not deform the sample enough. A typical resonance frequency should be sufficient to swing the cantilever in the liquid (i.e., be at least a few tens of kHz). The contact area should be small compared to the size of the cell wall since the sample under study in Young's modulus calculations is assumed to be an infinite half-space. The following cantilevers were used for rye and maize roots: sharp cantilevers with a typical resonance frequency of 60 kHz, an average spring constant of 1.5 N·m^-1, and an apex radius of 10 nm, or spherical cantilevers with a typical resonance frequency of 75 kHz, an average spring constant of 2.8 N·m^-1, and an apex radius of 150 nm.
2. Switch on the AFM device and the associated software (Table of Materials). Mount the cantilever on the tip holder for the liquid cell. Place a drop of liquid on the tip to avoid air bubbles forming on the tip while immersing it in the liquid.
   NOTE: In case of bubble formation, the liquid can be removed with a precision wipe, and then a new drop can be placed.
3. Mount the tip holder on the scanning head.
4. Place a hard sample (fresh glass slide) into a Petri dish cap. Pour the liquid into it, ensuring it covers the slide. Place the scanning head on the stage and raise the specimen so that the liquid covers the cantilever.
5. Choose the Heads tab in the drop-down Tools menu. Make sure that the scanning head for the liquid cell is selected.
6. Click on the Aiming button to open the Laser Aiming window. Click on the Camera button to open the Optical Microscope window. Position the laser at the tip of the cantilever, using the screws on the AFM head and the Optical Microscope window for orientation.
7. Choose the Semicontact mode from the drop-down menu in the main window of the program.
8. Open the Resonance tab and choose the type of cantilever being used in the Probes menu. Click on the Auto button to determine the resonance frequency.
   NOTE: If the cantilever being used is not listed, open Tools > Probe Passport. Create a new file, enter the cantilever parameters, and save it. Then choose it in the Probes drop-down menu.
9. Choose the Contact mode from the drop-down menu in the main window of the program.
10. Click on the N_Force Cal button to open cantilever calibration window. Choose the cantilever being used and click on the Sweep button, and then the Spect Meas button to determine the cantilever spring constant using the thermal-tune procedure. Close the window.
11. Set the SetPoint to 1 nA. Open the Approach tab and click on the Landing button to approach the sample.
12. Open the Scanning tab. Set the scan Rate to 0.5 Hz. Click on the Area button and set the scan Size to 10 µm x 10 µm and the scan Point to 256 x 256. Click on the Run button and scan about 20 lines to check that there is no contamination on the glass. Click on the Stop button to end the scan without losing data.
13. Open the Curves tab. Set parameters for retract and approach to 500 nm and -100 nm, respectively.
14. Choose the last scan in the drop-down menu Select Frame to observe the surface.
15. Find a clean area on the glass and indicate the point where the force-deformation curve should be taken, and click on the Run button to get the curve. Record three to five force curves in different places on the scan.
16. Click on the Data button to open the analysis window and select the frame with the force-deformation curve.
17. Click on the Pair Markers button and indicate the linear part of the retraction curve to calculate the ratio of the DFL signal to the displacement, which is the Inverse Optical Sensitivity (IOS, nA nm^-1) or deflection sensitivity.
18. Repeat step 2.17 for all recorded curves and write down all calculated values of the DFL signal to displacement ratio. They should be the same.
19. Close the analysis window, click on the Approach tab, and then the Remove button to retract the probe from the sample.

3. Data acquisition

1. Guide the sample under the AFM cantilever using the optical microscope. Click on the Approach tab, and then the Landing button to approach the sample in contact mode with a SetPoint of 1 nA.
2. Click on the Scanning tab, and then the Area button. Select the Area Size of 70 µm x 70 µm to scan.
3. Click on the Move Probe button and check the entire scan area by moving the scanner over it. Based on the degree of scanner protrusion, find the highest point.
4. Open the Approach tab, then click on the Remove button to retract from the sample. Using the highest point as a target, click on the Landing button to approach the sample again. Then, check the surface again by clicking on the Move Probe button and moving the scanner over it. None of the points in the area should require the scanner to be fully raised.
   NOTE: Ideally, the scanner z-range should exceed the height difference of the sample. However, it is acceptable if some points on the scan area are below the scanner's capacity. At the same time, there should not be many such points. Large areas which cannot be reached by the scanner may cause the scanning abortion.
5. Set the scan Rate to 0.5 Hz. Set the scan Size to 70 µm x 70 µm and the scan Point to 64 x 64. Click on the Run button and scan to check the surface of the sample and its possible contamination with agarose.
   NOTE: If the sample has cells with a large lumen, the size of the scan area can be reduced, or it should be moved so that there are more cell walls on the scan.
6. Click on the On button at the top of the main window to switch off the feedback loop.
7. Choose the HDPlus mode (force-volume method) in the drop-down menu in the main window of the program. An additional window (HD window) will appear.
8. Set the SetPoint to 0.1 nA in the main program window.
9. In the Main tab of the HD window, set the scanning parameters (amplitude and frequency of sinusoidal cantilever movement) appropriate to the studied sample.
   NOTE: Each type of sample and cantilever requires the selection of scanning parameters. Some preliminary experiments are necessary. For maize or rye roots, sharp and spherical cantilevers were used at an amplitude of 400 nm and a frequency of 300 Hz.
10. Open the Noises tab in the HD window and enter the resonance frequency of the cantilever.
11. Open the Quant tab of the HD window and enter the IOS, cantilever Stiffness, and tip radius and angle. Select the model of contact which will be used for calculations depending on tip geometry.
    NOTE: DMT model takes into account the adhesion force existing between the probe and the sample, and is most commonly used in mechanobiology²¹.
12. Open the Scan tab of the HD window and select the signals to be recorded. Select the direction in which the signal is recorded. Tick the Force volume box to get a record of all force curves.
    NOTE: Record the elasticity modulus signal in both forward and backward directions. It will provide more points to calculate.
13. Click on the Off button at the top of the main program window to switch on the feedback loop.
14. Click on the PhaseCorr button in the main HD window to correct the sensitivity of the optical system.
15. The vs Time tab of the main HD window presents the function of the DFL signal vs. time in real-time; select the parts of this function that will be used for baseline level determination and to fit the contact model for further calculations.
16. Set the scan Point value to 256 x 256 in the main window of the program. Set the scan Rate to 0.2 Hz and click on the Run button to scan the sample.
17. After scanning stops, click on the On button at the top of the main window to switch off the feedback loop.
18. Choose Contact mode in the drop-down menu, open the Approach tab, and click on the Remove button to retract from the sample.
19. Click on the Data button to open the analysis software and save the output.
20. At the end of the measurement day, remove the cantilever tip holder from the head and carefully rinse it with ultrapure water several times. Each time, remove the water with a precision wipe.

4. Data evaluation and post-processing

NOTE: Do not rely on elastic modulus values calculated automatically. Since the surface varies greatly in height, many artifact curves should be expelled.

1. Open the saved file in the analysis software.
2. Select the HDForceVolume frame.
3. Hold down the Ctrl key and select one visual frame obtained in the same direction of scanning. Click on the Load External Map button to see where the cell walls are located.
   NOTE: Any signal map can be used as a visual frame, but using the DFL signal map (different instruments may refer it to as a deflection signal or error signal) may be the easiest way.
4. Check InvOptSens (IOS) and cantilever Stiffness values in the Main tab.
5. Open the Additional tab and check the tip parameters and contact model.
6. Click on various points on the cell walls on the visual frame and select only those curves that are well described by the model. See Representative Results for details.
7. Write down the obtained values.

Representative Results

Typical elastic modulus and DFL maps, as well as force curves obtained on rye and maize roots by the method described, are presented in Figure 2. Figure 2A shows elastic modulus and DFL maps obtained on the transverse section of rye primary root. The white areas in the modulus map (Figure 2A, left) correspond to an erroneous overestimation of Young's modulus due to the scanner reaching its limit in the z-direction. This image is not convenient to be used as an external map for a further selection of satisfactory force curves. The DFL signal map (deflection/error map) that is presented on the right (Figure 2A) is better suited for this purpose.

The modulus maps of transverse and longitudinal sections of the maize root are shown in Figure 2B,C. The scanner that has been fully extended while trying to reach the bottom of the sample can result in erroneous measurements and even interrupted scans (top part of Figure 2C). Sometimes, after getting a good scan in contact mode, the instrument suddenly contracts the scanner completely, alarming that it has reached the maximum z-direction limit when it switches to the force-volume mode. It means that the sample was not immobilized properly and floated up. A new sample should be prepared.

The agarose contamination may also be a reason to prepare a new sample (Figure 2D,E). The agarose presence may be checked while obtaining the first scan in the contact mode (step 3.5). Ideally, the cell walls and some cell bottoms should be visible on such scans (Figure 2D); however, in case of inaccurate immobilization, the surface may be covered with agarose (Figure 2E), which masks the sample topography.

Figure 2F shows four different force curves recorded at different points of the same cell wall. The curve recorded at Point 1 shows no baseline. It means there was no separation of the cantilever tip from the cell wall. The modulus calculated from this curve was overestimated. The curve recorded at Point 3 shows a shoulder on the approaching part. This artifact may indicate that the cell wall was bent. Both curves recorded at Points 1 and 3 should be expelled. Points 2 and 4 show satisfactory force curves with similar values of moduli. The following criteria can be used to distinguish proper curves: (i) the baseline level should be evident on both approaching and retracting parts of the curve; (ii) there should be no irregularities such as shoulders or failures (appears as constant force values and a sudden decrease of force values in the middle of the slope, respectively); (iii) the selected model should fit the curve well. For all selected curves, the maximum applied force values should be similar. A maximum applied force that is significantly higher or lower than the expected value can also be used as a criterion for curve expelling. Using neural networks to discard poor-quality curves¹⁹ can speed up the processing of results.

Figure 1: Critical steps in sample preparation. (A) Root fragment embedded in a block of agarose and mounted on the vibratome stage. (B) Sample section placed on top of a layer of 1% agarose and immobilized with additional agarose in a Petri dish cap. Please click here to view a larger version of this figure.

Figure 2: Typical AFM images and measurements on primary roots. (A) Elasticity modulus and DFL maps of a transverse section of the rye root. Fully recorded (B) and interrupted (C) maps of elasticity modulus obtained on longitudinal and transverse sections of maize root. 3D view of clean (D) and agarose-covered (E) sections of maize roots. (F) A DFL map of a transverse section of maize root. Points 1-4 on the cell wall show the positions where the force curves presented on the right side of the image were recorded. The red line is for approaching, the blue line is for retraction, and the black line shows the contact mechanics model fit. The curves recorded at points 1 and 3 have artifacts and cannot be used for calculations. Rye roots were examined with sharp tips, and maize roots were examined with spherical tips. Abbreviations: Rhiz = rhizodermis, Exo = exodermis, OC = outer cortex, IC = inner cortex, End = endodermis, Per = pericycle, and VP = vascular parenchyma. Scale bars = 10 µm. Please click here to view a larger version of this figure.

Discussion

The mechanical properties of the primary cell walls determine the direction and rate of plant cell growth, and therefore the future size and shape of the plant. The AFM-based method presented here complements existing techniques which are used to study the properties of plant cell walls. It allows the elasticity of cell walls, which belong to the inner tissues of the plant, to be investigated. Using the presented method, the mechanical properties of cell walls in different tissues of the growing maize root were mapped, and a mathematical model was constructed based on these properties⁶. The simulated root responded to the application of pressure imitating turgor by changes in linear size. These changes were in the same range and in the opposite direction as those exhibited by a freshly excised root in plasmolyzing medium⁶.

The sample preparation is critical for this protocol. The drying of the sample should be avoided during the entire preparation. Correct placement of the plant material to be sectioned is an important issue; the cell walls must be oriented strictly perpendicular to the future scanning plane. Otherwise, unsatisfactory force-deformation curves will be obtained because the cell wall has been bent or buckled. The agarose-embedded section should be tightly fixed with additional agarose (Figure 2B). The latter should not get onto the sample, otherwise, the agarose properties will be measured. The most important part of this protocol is to filter the resulting force curves. Even studying model objects, such as AFM calibration grids, may result in a significant proportion of erroneous force-deformation curves²². Direct use of the calculated modulus values without analyzing the underlying force-deformation curves can lead to incorrect conclusions (Figure 2F).

The technique is not limited by the localization of the cell or the size of plant organs. However, objects that are too small are difficult to cut, orient, and immobilize. Cells with a large lumen are difficult to work with since there is a high probability of scan interruption (Figure 2C).

With all the benefits of using non-fixed plant sections prepared with a vibratome, there are at least two crucial issues: the possibility of developing a stress reaction after cutting plant material, and the possibility of partial solubilization of material from the cut cell walls. Both can lead to changes in the composition and properties of cell walls. The presence of immediate stress response on the cut was tested on Arabidopsis apices¹⁰. No significant effect on the observed properties was reported. Experiments were conducted to test changes in the properties of cut cell walls during their incubation in water¹⁹. There was no stable trend in apparent Young's modulus values measured on the same wall of maize root for at least a half-hour period¹⁹. However, it is better to maintain the time of experiments as short as possible and check each specimen for such changes during incubation of the sample in water.

AFM is a powerful technique for studying plant cell walls. However, it is obvious that calculations of elasticity modulus, stiffness, and adhesion are based on several assumptions that are not fulfilled by plant cell walls. All models of contact mechanics used by AFM consider the investigated specimen as an infinite half-space of isotropic material, which is not true for plant cell walls. Simultaneously, it is assumed that the geometry and properties of the indenter are accurately measured and constant over the cantilever lifespan, which may also be erroneous. In addition, plant cell walls exhibit complex mechanical behavior involving elastic, viscoelastic, and plastic components, and usually only one of these can be measured in a single experiment. Nevertheless, the data obtained using AFM-based methods reliably correlate with the plant cell growth and mechanical performance in various organs and species^3,4,5,6. All this means that both the technique itself and the understanding of the data obtained with it can be improved.

Materials

Name	Company	Catalog Number	Comments
Agarose, low melting point	Helicon	B-5000-0.1	for sample fixation
Brush	-	-	for section moving
Cantilevers	NanoTools, Germany	NT_B150_v0020-5	Model: Biosphere B150-FM
Cantilevers	NT-MDT, Russia	FMG01/50	Model: FMG01
Cyanoacrylate adhesive	-	-	for vibratomy
Glass slides	Heinz Herenz	1042000	for vibratomy and AFM calibration
ImageAnalysis P9 Software	NT-MDT, Russia	-	for data analysis
Leica DM1000 epifluorescence microscope	Leica Biosystems, Germany	11591301	for section check
NaOCl	-	-	for seed sterilization
Nova PX 3.4.1 Software	NT-MDT, Russia	-	for experiments conducting
NTEGRA Prima microscope with HD controller	NT-MDT, Russia	-	for AFM and data acquisition
Petri dish 35 mm	Thermo Fisher Scientific	153066	for sample fixation
Tip pipette 1000 µL	Thermo Fisher Scientific	4642092	-
Tip pipette 2-20 µL	Thermo Fisher Scientific	4642062	-
Ultrapure water	-	-	-
Vibratome Leica VT 1000S	Leica Biosystems, Germany	1404723512	for sample sectioning